Upload folder using huggingface_hub

checkpoint-200/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a70a9af242ff0e1565305fd79d6347e3d0ab897fd3729c7b65171ebde4ca5b9
+oid sha256:f516c96875369a56b3695683bcaeeda6eb2a60e651928b627dc327f2c1940c37
 size 966995080

checkpoint-200/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a2245dda281cf7f0034c402bf28a80422c2ea2616455d75ee7003f5ea2117d9
+oid sha256:37abe716a191e7469cc08de7a38a4975038c5d4ae5393c33e7fc18ca427984c7
 size 1925064044

checkpoint-200/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9bd55b3848d82967a207e0805911c79200c6adce71e3b37fd24549a718f75738
+oid sha256:2412050386e127bb591a2ab33de2372b85e96145ca2326e6c8e125dbabd1d804
 size 988

checkpoint-200/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:031e0fdc2c20f6db35d37f449b7cbb318d204bbc0f5803ee3adfbcbdb6426566
+oid sha256:c805b003744d6c00853d1b84790e23f138d0a52d5331bb5dfe3ab6b5b38c6240
 size 1064

checkpoint-200/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 200,
-  "best_metric": 6.366537880060474,
+  "best_metric": 10.44776119402985,
   "best_model_checkpoint": "./SALAMA_C7/checkpoint-200",
   "epoch": 2.0833333333333335,
   "eval_steps": 200,
@@ -11,88 +11,88 @@
   "log_history": [
     {
       "epoch": 0.20833333333333334,
-      "grad_norm": 2.567998170852661,
-      "learning_rate": 1.9000000000000002e-06,
-      "loss": 0.1774,
+      "grad_norm": 11.919928550720215,
+      "learning_rate": 1.8000000000000001e-06,
+      "loss": 0.6563,
       "step": 20
     },
     {
       "epoch": 0.4166666666666667,
-      "grad_norm": 3.8020272254943848,
-      "learning_rate": 3.900000000000001e-06,
-      "loss": 0.1662,
+      "grad_norm": 11.068168640136719,
+      "learning_rate": 3.8000000000000005e-06,
+      "loss": 0.5814,
       "step": 40
     },
     {
       "epoch": 0.625,
-      "grad_norm": 2.032432794570923,
-      "learning_rate": 5.9e-06,
-      "loss": 0.1198,
+      "grad_norm": 7.94372034072876,
+      "learning_rate": 5.8e-06,
+      "loss": 0.5414,
       "step": 60
     },
     {
       "epoch": 0.8333333333333334,
-      "grad_norm": 1.15700101852417,
-      "learning_rate": 7.9e-06,
-      "loss": 0.109,
+      "grad_norm": 6.410305500030518,
+      "learning_rate": 7.800000000000002e-06,
+      "loss": 0.4745,
       "step": 80
     },
     {
       "epoch": 1.0416666666666667,
-      "grad_norm": 3.478109121322632,
-      "learning_rate": 9.9e-06,
-      "loss": 0.0838,
+      "grad_norm": 4.956066131591797,
+      "learning_rate": 9.800000000000001e-06,
+      "loss": 0.3529,
       "step": 100
     },
     {
       "epoch": 1.25,
-      "grad_norm": 1.5184258222579956,
-      "learning_rate": 9.779069767441862e-06,
-      "loss": 0.0669,
+      "grad_norm": 3.679781436920166,
+      "learning_rate": 9.042553191489362e-06,
+      "loss": 0.2865,
       "step": 120
     },
     {
       "epoch": 1.4583333333333333,
-      "grad_norm": 4.6623759269714355,
-      "learning_rate": 9.546511627906978e-06,
-      "loss": 0.0833,
+      "grad_norm": 3.3053739070892334,
+      "learning_rate": 7.97872340425532e-06,
+      "loss": 0.263,
       "step": 140
     },
     {
       "epoch": 1.6666666666666665,
-      "grad_norm": 3.8098835945129395,
-      "learning_rate": 9.313953488372095e-06,
-      "loss": 0.0915,
+      "grad_norm": 5.734177589416504,
+      "learning_rate": 6.914893617021278e-06,
+      "loss": 0.2875,
       "step": 160
     },
     {
       "epoch": 1.875,
-      "grad_norm": 2.154282331466675,
-      "learning_rate": 9.08139534883721e-06,
-      "loss": 0.0876,
+      "grad_norm": 7.049015522003174,
+      "learning_rate": 5.851063829787235e-06,
+      "loss": 0.2821,
       "step": 180
     },
     {
       "epoch": 2.0833333333333335,
-      "grad_norm": 2.825495719909668,
-      "learning_rate": 8.848837209302326e-06,
-      "loss": 0.1009,
+      "grad_norm": 2.211012363433838,
+      "learning_rate": 4.787234042553192e-06,
+      "loss": 0.204,
       "step": 200
     },
     {
       "epoch": 2.0833333333333335,
-      "eval_loss": 0.08810210227966309,
-      "eval_runtime": 266.2349,
-      "eval_samples_per_second": 2.881,
-      "eval_steps_per_second": 0.361,
-      "eval_wer": 6.366537880060474,
+      "eval_loss": 0.18036405742168427,
+      "eval_runtime": 274.0379,
+      "eval_samples_per_second": 2.799,
+      "eval_steps_per_second": 0.35,
+      "eval_wer": 10.44776119402985,
       "step": 200
     }
   ],
   "logging_steps": 20,
-  "max_steps": 960,
+  "max_steps": 288,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 3,
   "save_steps": 200,
   "stateful_callbacks": {
     "TrainerControl": {

checkpoint-200/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0fe98a9ee4e1bd23d9bf7869649307b5338a198e11e58c7d70a840df6d10e0f6
+oid sha256:ac55cd8bbcced756202be3aca523f94ecd6548025e9d1a1609129703b17c0d6c
 size 5496

checkpoint-288/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0459fdd50caeddc1dddcfe48a68ce9b966d455fa8c104fe166f29b4a007f1c1c
+oid sha256:3487346ccfeb08676834c610a07f3a5bbeedb50749a2e7fb0a8fba167c41b2f4
 size 966995080

checkpoint-288/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31a2cee16bc237b2d704ceea75812d061144601bb291c3deb683ceb9cb6c3c70
+oid sha256:88dd23a6795ecc80ab0f1f87e68d52393c275ef3d4b6a28f74de7e9d0c68a663
 size 1925064044

checkpoint-288/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef86260123ed82be82b42a2f4de267130a1b8d733ee3522c344257175f57e900
+oid sha256:def1e6b141b8d9b6306daa1675d0c5192ba3a9a3dba69e9798a62bdc7129b459
 size 988

checkpoint-288/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 200,
-  "best_metric": 9.978162271123804,
+  "best_metric": 10.44776119402985,
   "best_model_checkpoint": "./SALAMA_C7/checkpoint-200",
   "epoch": 3.0,
   "eval_steps": 200,
@@ -11,109 +11,109 @@
   "log_history": [
     {
       "epoch": 0.20833333333333334,
-      "grad_norm": 10.547709465026855,
+      "grad_norm": 11.919928550720215,
       "learning_rate": 1.8000000000000001e-06,
-      "loss": 0.7281,
+      "loss": 0.6563,
       "step": 20
     },
     {
       "epoch": 0.4166666666666667,
-      "grad_norm": 11.514107704162598,
+      "grad_norm": 11.068168640136719,
       "learning_rate": 3.8000000000000005e-06,
-      "loss": 0.6221,
+      "loss": 0.5814,
       "step": 40
     },
     {
       "epoch": 0.625,
-      "grad_norm": 7.552187442779541,
+      "grad_norm": 7.94372034072876,
       "learning_rate": 5.8e-06,
-      "loss": 0.4623,
+      "loss": 0.5414,
       "step": 60
     },
     {
       "epoch": 0.8333333333333334,
-      "grad_norm": 5.789488315582275,
+      "grad_norm": 6.410305500030518,
       "learning_rate": 7.800000000000002e-06,
-      "loss": 0.4368,
+      "loss": 0.4745,
       "step": 80
     },
     {
       "epoch": 1.0416666666666667,
-      "grad_norm": 5.5591630935668945,
+      "grad_norm": 4.956066131591797,
       "learning_rate": 9.800000000000001e-06,
-      "loss": 0.3215,
+      "loss": 0.3529,
       "step": 100
     },
     {
       "epoch": 1.25,
-      "grad_norm": 3.7926101684570312,
+      "grad_norm": 3.679781436920166,
       "learning_rate": 9.042553191489362e-06,
-      "loss": 0.2535,
+      "loss": 0.2865,
       "step": 120
     },
     {
       "epoch": 1.4583333333333333,
-      "grad_norm": 6.078831672668457,
+      "grad_norm": 3.3053739070892334,
       "learning_rate": 7.97872340425532e-06,
-      "loss": 0.2608,
+      "loss": 0.263,
       "step": 140
     },
     {
       "epoch": 1.6666666666666665,
-      "grad_norm": 5.2472662925720215,
+      "grad_norm": 5.734177589416504,
       "learning_rate": 6.914893617021278e-06,
-      "loss": 0.2671,
+      "loss": 0.2875,
       "step": 160
     },
     {
       "epoch": 1.875,
-      "grad_norm": 3.961386203765869,
+      "grad_norm": 7.049015522003174,
       "learning_rate": 5.851063829787235e-06,
-      "loss": 0.2477,
+      "loss": 0.2821,
       "step": 180
     },
     {
       "epoch": 2.0833333333333335,
-      "grad_norm": 3.207435369491577,
+      "grad_norm": 2.211012363433838,
       "learning_rate": 4.787234042553192e-06,
-      "loss": 0.244,
+      "loss": 0.204,
       "step": 200
     },
     {
       "epoch": 2.0833333333333335,
-      "eval_loss": 0.15180820226669312,
-      "eval_runtime": 341.1778,
-      "eval_samples_per_second": 2.248,
-      "eval_steps_per_second": 0.281,
-      "eval_wer": 9.978162271123804,
+      "eval_loss": 0.18036405742168427,
+      "eval_runtime": 274.0379,
+      "eval_samples_per_second": 2.799,
+      "eval_steps_per_second": 0.35,
+      "eval_wer": 10.44776119402985,
       "step": 200
     },
     {
       "epoch": 2.2916666666666665,
-      "grad_norm": 3.7495334148406982,
+      "grad_norm": 4.120218276977539,
       "learning_rate": 3.723404255319149e-06,
-      "loss": 0.1436,
+      "loss": 0.1378,
       "step": 220
     },
     {
       "epoch": 2.5,
-      "grad_norm": 4.28264856338501,
+      "grad_norm": 2.3007442951202393,
       "learning_rate": 2.6595744680851065e-06,
-      "loss": 0.152,
+      "loss": 0.1111,
       "step": 240
     },
     {
       "epoch": 2.7083333333333335,
-      "grad_norm": 2.525714159011841,
+      "grad_norm": 4.804255485534668,
       "learning_rate": 1.595744680851064e-06,
-      "loss": 0.1231,
+      "loss": 0.1654,
       "step": 260
     },
     {
       "epoch": 2.9166666666666665,
-      "grad_norm": 2.2750539779663086,
+      "grad_norm": 3.7491042613983154,
       "learning_rate": 5.319148936170213e-07,
-      "loss": 0.116,
+      "loss": 0.1346,
       "step": 280
     }
   ],

checkpoint-288/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3ffc78527769250dfdd03630f788573d15e547f1e38bf61f7517e9ee58ba54e
+oid sha256:ac55cd8bbcced756202be3aca523f94ecd6548025e9d1a1609129703b17c0d6c
 size 5496